DE(1)
DE(1)
AbEC - DE
> [ Home ] > [ Examples ]
[~]$ The algorithm

The Differential Evolution (DE) algorithm is a population-based optimization algorithm that was proposed by Rainer Storn and Kenneth Price in 1997. It is commonly used to solve optimization problems, particularly in continuous domains.
The DE algorithm is based on the concept of evolutionary computation, inspired by the process of natural selection. It iteratively improves a population of candidate solutions to find the optimal solution. The algorithm operates on a population of individuals, where each individual represents a potential solution to the optimization problem.

If you want to know more about it, please come here. Below the configuration files to run the DE using the framework.

[~]$ Configuration files

> algoConfig.ini

> [ file ]

{
"__COMMENT__": "BASIC CONFIGURATION",
"ALGORITHM": "DE", -> Configure the name
"POPSIZE": 50,-> Set the population size
"MIN_POS": 0,-> Set the min value in the search space
"MAX_POS": 100,-> Set the max value in the search space
"__COMMENT__": "OPTIMIZER CONFIGURATION",
"GA_POP_PERC": 0,
"GA_ELI_PERC": 0.2,
"GA_CROSS_PERC": 1,
"GA_MUT_PERC": 0.1,
"GA_MUT_STD": 1,
"GA_ENCODER": 0,
"GA_INDSIZE": 16,
"PSO_POP_PERC": 0,
"PSO_PHI1": 2.05,
"PSO_PHI2": 2.05,
"PSO_W": 0.729,
"PSO_MIN_VEL": -100,
"PSO_MAX_VEL": 100,
"DE_POP_PERC": 1,-> Set the whole population to perform DE
"DE_F": 0.5,-> Set the F hyper-parameter in 0.5
"DE_CR": 0.7,-> Set the CR hyper-parameter in 0.7
"ES_POP_PERC": 0,
"ES_RCLOUD": 0.2,
"__COMMENT__": "COMPONENTS CONFIGURATION",
"COMP_CHANGE_DETECT": 0,
"COMP_CHANGE_DETECT_MODE": 0,
"COMP_MULTIPOP": 0,
"COMP_MULTIPOP_N": 10,
"COMP_MUT": 0,
"COMP_MUT_PERC": 0.05,
"COMP_MUT_ELI": 0.5,
"COMP_MUT_STD": 0.1,
"COMP_EXCLUSION": 0,
"COMP_EXCLUSION_REXCL": 22.9,
"COMP_ANTI_CONVERGENCE": 0,
"COMP_ANTI_CONVERGENCE_RCONV": 39.7,
"COMP_LOCAL_SEARCH": 0,
"COMP_LOCAL_SEARCH_ETRY": 20,
"COMP_LOCAL_SEARCH_RLS": 1
}

Be free to use this framework.
AbEC © 2023.